US20030110457A1 - Method and program product for designing hierarchical circuit for quiescent current testing and circuit produced thereby - Google Patents
Method and program product for designing hierarchical circuit for quiescent current testing and circuit produced thereby Download PDFInfo
- Publication number
- US20030110457A1 US20030110457A1 US10/011,128 US1112801A US2003110457A1 US 20030110457 A1 US20030110457 A1 US 20030110457A1 US 1112801 A US1112801 A US 1112801A US 2003110457 A1 US2003110457 A1 US 2003110457A1
- Authority
- US
- United States
- Prior art keywords
- block
- memory elements
- circuit
- blocks
- quiescent current
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/28—Testing of electronic circuits, e.g. by signal tracer
- G01R31/30—Marginal testing, e.g. by varying supply voltage
- G01R31/3004—Current or voltage test
- G01R31/3012—Built-In-Current test [BIC]
Definitions
- the present invention relates, in general to the design and testing of integrated circuits and, more specifically, to a method and a program product for designing circuits in a manner to facilitate quiescent current (IDDq) testing of the circuit and to a circuit designed according to the method.
- IDDq quiescent current
- design hierarchy The complexity of semiconductor circuits often requires partitioning the circuit design into several core modules or blocks, generally referred to as design hierarchy.
- the blocks can be verified and laid out independently from each other.
- This design approach has been used for a long time and has several benefits, the most significant of which are to accelerate the development of the design by allowing several designers to work in parallel and to reduce the difficulty for design automation tools by avoiding the need to manipulate the description of the entire circuit at once.
- scan test methods can be adapted to a hierarchical design methodology by adding scannable memory elements around the periphery of core modules or by modifying functional memory elements near the periphery of modules and providing an internal and external mode of operation to these memory elements (see, for example, “A structured and Scalable Mechanism for Test Access to Embedded Reusable Cores”, ITC '98. paper 12. 1, Marinessen et al. and Applicants' co-pending U.S. patent application Ser. No. 09/626,877 filed on Jul.
- the preparation of the scan tests involves at least the verification of design-for-test rules and calculation of test coverage.
- the preparation also involves the generation of a simplified model of the module that can be used to perform the same steps at the next level of hierarchy of the design.
- the use of the simplified model eliminates the need to manipulate the entire description of core modules.
- the state of memory elements that are not involved i.e. not near the periphery of the module
- Core modules can be tested in parallel or sequentially.
- the present invention provides a hierarchical design method for integrated circuits to be tested using quiescent current monitoring and a circuit designed according to the method.
- the method allows the state of the entire circuit to be set in a repeatable and predictable manner which avoids the need to analyze the entire circuit at once.
- the circuit is divided into hierarchical blocks or modules according to a design hierarchy.
- the blocks have scannable memory elements which are organized into scan chain segments with at least one segment being a peripheral scan chain segment.
- the peripheral segment is comprised of input and output peripheral scannable memory elements which are configurable in an internal test mode and an external test mode.
- One aspect of the present invention is broadly defined as a method of designing integrated circuits, having at least one hierarchical block, for quiescent current testing, comprising analyzing each of one or more selected hierarchical blocks independently of other selected blocks in the same or higher levels of hierarchy to identify circuit states of each the selected blocks which could result in elevated quiescent current levels during quiescent current testing of the circuit, the analyzing beginning with blocks at a lowest level of design hierarchy and proceeding in sequence through each level of design hierarchy to a highest level of hierarchy containing a top-level block; and calculating a fault coverage of for each the selected block.
- Another aspect of the present invention is generally defined as a circuit having at least one hierarchical block, the circuit comprising: each block having scannable memory elements arranged into two or more scan chain segments having a serial input and a serial output, the memory elements including peripheral memory elements associated with block inputs and outputs, the peripheral scannable memory elements being arranged in peripheral scan chain segments, input peripheral elements being configurable in internal test mode for testing circuitry within the block and output peripheral memory elements being configurable in external test mode for testing circuitry external of the block; means associated with each segment and responsive to an active quiescent current test signal for connecting the serial input of the segment to a block serial input, or to the serial output of another scan chain segment, or to the serial output of a peripheral scan chain segment of a block located at one level of hierarchy down; and means responsive to the active quiescent current test signal for concurrently configuring input peripheral memory elements of the block in internal test mode and configuring output peripheral memory elements of the block in an external test mode during quiescent current testing of the circuit.
- the layout of the circuit does not need to be hierarchical in order to apply the proposed hierarchical quiescent current test method of the present invention. However, if the layout of the circuit is hierarchical, the design hierarchy used for the layout will typically correspond to the top levels of the design hierarchy used for the test method.
- Still another aspect of the present invention is defined as a method of performing quiescent current testing on an integrated circuit having at least one hierarchical logic block, each block having a block serial input, scannable memory elements arranged in at least one scan chain having a scan chain serial input, the method comprising connecting each block serial input to a circuit serial input and connecting each scan chain serial input to its associated block serial input; serially loading a quiescent current test pattern through the circuit serial input so as to load the test pattern into each scan chain in parallel until a predetermined reference point of the test pattern has been reached at which all scannable memory elements in the circuit have been initialized with a known value, and then performing a predetermined number of quiescent current measurement steps including: measuring the quiescent current of the circuit, storing the measured quiescent current value and the test pattern point at which the value was measured; and, while the predetermined number of measurements has not been reached, scanning in an additional sequence of bits of the test pattern and repeating the measurement steps.
- FIG. 1 is a diagrammatic block view of a circuit according to one embodiment of the present invention.
- FIG. 2 is a diagrammatic view of a hierarchical block showing scannable memory elements arranged in scan chains and peripheral memory elements associated with block inputs and outputs;
- FIG. 3 diagrammatically illustrates a prior art scannable memory element
- FIG. 4 diagrammatically illustrates a circuit having three levels of hierarchy and showing the manner in which scan chain segments may be connected according to one embodiment of the present invention
- FIG. 5 is similar to FIG. 4, but shows a more restrictive manner in which scan chain segments may be connected according to a preferred embodiment of the present invention
- FIG. 6 is a flow chart illustrating the method of performing a quiescent current test on a circuit designed according to an embodiment of the method of design of the present invention
- FIG. 7 is a flow chart generally illustrating a method of designing a hierarchical circuit for quiescent current testing according to an embodiment of the present invention
- FIG. 8 is a diagram illustrating a single scan chain configuration for two hierarchical blocks located one level down in design hierarchy from a parent or host block;
- FIG. 9 is a diagram similar to FIG. 8, but illustrating a parent block scan chain configuration which includes all scannable memory elements of the parent or top-level block and the peripheral memory elements of two hierarchical block located one level down in design hierarchy;
- FIG. 10 is a diagram similar to FIG. 8, but illustrating a single scan chain configuration for each of a host block and two hierarchical blocks located one level down in design hierarchy, which is used when performing a quiescent current test of the circuit;
- FIG. 11 diagrammatically illustrates the alignment of the serial input of three scan chain segments receiving the same quiescent current test pattern from a pseudo-random pattern generator
- FIG. 12 is a state diagram of a Test Access Port according to the IEEE 1149.1 standard.
- integrated circuit devices are typically designed and fabricated using one or more computer data files that define the layout of the circuit arrangements of the devices.
- the files are typically generated by a design tool and are subsequently used during manufacturing to create the layout masks that define the circuit arrangements applied to a semiconductor wafer.
- the programs are provided in a predefined format using a hardware description language (HDL) such as VHDL, verilog, EDIF, etc.
- HDL hardware description language
- signal bearing media include but are not limited to recordable type media such as volatile and non-volatile memory devices, floppy disks, hard disk drives, CD-ROM's, and DVD's, among others and transmission type media such as digital and analog communications links.
- FIG. 1 illustrates a simple hierarchical circuit 10 according to one embodiment of the present invention.
- the circuit includes a top-level hierarchical block 12 , two embedded hierarchical functional blocks 14 and 16 , a boundary scan register 18 , a Test Access Port (TAP) 20 , and a test controller 22 .
- Reference numeral 24 represents logic in top-level block 12 .
- Blocks 14 and 16 are contained within or embedded in the top-level block and, thus, are located one level down in design hierarchy. Two hierarchical blocks are shown for illustration purposes only. It is to be understood that the circuit may have any arbitrary number of hierarchical blocks and levels.
- the TAP and boundary scan register are preferred, but not essential to the invention.
- blocks 14 and 16 provide a unique function.
- FIG. 2 is a diagrammatic view of the internal structure of a typical hierarchical block.
- the block generally includes combinational logic 30 and block inputs 32 and block outputs 34 , by which the block communicates with other blocks during normal operation of the circuit, and memory or storage elements 36 which store data during normal operation of the circuit as well as during various test modes of operation.
- the memory elements are designed to be scannable to allow a test stimulus to be scanned into the elements and a test response data to be scanned out.
- FIG. 3 illustrates one arrangement of a typical scannable memory element 36 .
- the element receives input from combinational logic 30 .
- the memory element includes a flip-flop 38 having a data input, D, a data output Q and a clock input, CK.
- Input D is connected to the output of a multiplexer 40 .
- One input 42 of the multiplexer is connected to a memory element serial input, SI.
- the other input 44 of the multiplexer is connected to an output of combinational logic 30 .
- a control input 46 of the multiplexer receives a scan enable signal, SE, which operates to configure the memory element in scan or shift mode when SE is active (logic 1) and in capture or normal mode when SE is inactive (logic 0).
- Output Q also serves as the memory element Serial Output, SO.
- the scannable memory elements are serially interconnected into one or more scan chains 48 by connecting the serial output SO of each memory element to the serial input SI of another memory element.
- the serial input of the first memory element in the chain is connected to a scan chain serial input.
- the serial output of the last memory element in a scan chain is connected to a scan chain serial output.
- a scan chain segment is comprised of one or more memory elements.
- peripheral memory elements Memory elements located at the periphery or boundary of a logic block are referred herein to as “peripheral memory elements”, but also known in the art as “partitioning scan cells”. Elements which output a signal from the block, such as memory elements 50 , are output peripheral memory elements. Elements which receive a signal applied to the block, such as memory elements 52 , are input peripheral memory elements.
- the preferred form of peripheral memory elements are of the type disclosed and claimed in Applicant's above mentioned co-pending U.S. application patent application Ser. No. 09/626,877 which uses functional memory elements to partition a block into an internal partition and an external partition and thereby provide module or block isolation. However, the present invention can be employed in a circuit which uses conventional test dedicated peripheral memory elements. Both forms of peripheral elements are described and illustrated in the aforementioned application.
- Output peripheral memory elements are configurable in “external test mode” in which data contained in the memory element is applied to its associated block output node which, in turn, is connected to external circuitry. When the element is not in external test mode, data generated by the block is passed through to the output node. This allows test data to be scanned in to the output peripheral memory elements to test the response of external circuitry to various values of block outputs.
- input peripheral memory elements are configurable in “internal test mode” in which data contained in the peripheral memory element is passed on to the internal circuitry of the block. When the element is not in internal test mode, data applied to its associated block input node is applied to the internal circuitry of the block. This allows test data to be scanned in to the input peripheral memory elements to test the internal logic of the block.
- the top-level block need not necessarily be provided with peripheral memory elements.
- the boundary scan register chain can be used as a peripheral chain. However, this is not necessary because access to chip inputs and outputs is available.
- Peripheral memory elements may be included in the top-level block if the complete circuit may itself constitute a block which is embedded into another circuit.
- Input and output peripheral memory elements are normally not active concurrently.
- the test controller is provided with circuitry (not shown) which responds to an active quiescent current test signal by applying a signal to input and output peripheral memory elements which concurrently configures input peripheral memory elements in internal test mode and output peripheral memory elements in external test mode.
- This configuration is referred to herein as a “peripheral memory element quiescent current test mode”.
- This mode allows the peripheral memory elements to be arranged in a scan chain segment and a simplified model of embedded blocks to be substituted for the complete description of the block during analysis of a parent block, as explained more fully later in the description of the method of the present invention.
- test data scan in to input and output peripheral memory elements to be concurrently applied to block internal logic as well as to logic external to a block.
- One aspect of the present invention relates to the manner in which internal and peripheral memory elements are arranged into scan chains for use in loading quiescent current test patterns into the various hierarchical blocks in the circuit.
- memory element segments of a block are arranged, in IDDq test mode, on a scan path which originates from or includes a block serial input. This is achieved by connecting segment serial input to: A) a block input, B) the serial output of another segment or C) the serial output of a segment located in an embedded block one level down in design hierarchy and by connecting the block serial input of embedded blocks to the block serial input
- a segment comprises one or more serially configurable scannable memory elements.
- This configuration allows the use of a simplified or shell model of a block during a block analysis design phase, described later.
- the simplified model contains peripheral memory element segments and peripheral logic and additional files which describe the scan paths of the block and the scan paths of embedded blocks and allows for pattern generation.
- some rules checking e.g. scan path continuity
- verification/simulation is performed in an indirect manner in which other scan test modes are rules checked and verified/simulated and it assumed that the IDDq test mode is correct by construction.
- this general configuration does not place any restrictions on internal and peripheral segments.
- This configuration also allows the use of simplified models which contain all scan paths and peripheral logic which, in turn, allows for a complete analysis including rules checking, pattern generation and verification (i.e. simulation).
- the configuration provides significant size reduction in the size of simplified models, but is not optimal.
- FIG. 4 illustrates a circuit having three levels of hierarchy.
- the figure shows three blocks including a top level block 60 having an embedded block 62 which defines a second level of hierarchy and which, in turn, has an embedded block 64 which defines a third level of hierarchy.
- Scan chain segments are shown as a sequence of rectangles which represent memory elements. Peripheral segments are bolded. A segment may consist of one or more scannable memory elements.
- Blocks 60 , 62 and 64 are provided with block serial inputs 66 , 68 and 70 , respectively. For simplicity, each block is shown as having only one block serial input. However, more than one block serial input may be provided without departing from the spirit of the invention.
- Host block 60 includes two internal segments 72 and 74 and no peripheral segments. Segment 72 is connected to top level block serial input 66 , the chip circuit input.
- Embedded block 62 includes two internal segments 76 and 80 and a peripheral segment 78 .
- block 62 is provided with means, in the form of a multiplexer 82 , for selectively connecting the serial input of peripheral segment 78 to one of two sources including the output of internal segment 76 of block 62 or to the serial output of internal segment 72 of parent block 60 .
- Multiplexer 82 is controlled by a control signal Cl.
- the output of peripheral segment 78 is connected to the input of internal segment 74 of parent block 60 .
- Embedded block 64 which resides in the third level of hierarchy, has four segments, including internal segments 84 , 86 and 88 , and peripheral segment 90 . Segments 84 and 86 are serially connected together, with the input of segment 84 being connected to block serial input 70 .
- a multiplexer 92 receives the serial output of peripheral segment 78 in parent block 62 and the output of internal segment 86 and applies its output to peripheral segment 90 . Multiplexer 92 is controlled by control signal C 2 .
- the serial output of peripheral segment 90 branches to internal segment 80 of parent block 62 and internal segment 88 of current block 64 . It will be seen that all segments reside on scan path which originates from or includes a block serial input.
- Control signals C 1 and C 2 are mode signals which are used to define different scan chains configurations for scan testing and quiescent current testing of the circuit. A specific example of various modes is given below with reference to FIG. 5.
- IDDq test mode internal segments are arranged on a scan path which originates from or includes a block serial input by connecting the internal chain segment serial input to: A) a block input, B) the serial output of another segment (internal or peripheral) in the block or C) the serial output of a peripheral segment located in an embedded block one level down in design hierarchy.
- option (C) More specifically, if an internal segment is connected to any segment in an embedded block, it must be connected to the output of a peripheral segment.
- peripheral segments are arranged on a scan path which originates from or includes a block serial input by connecting the peripheral segment serial input to: A) a block input, B) the serial output of another peripheral segment of the block or C) the serial output of a peripheral segment located in an embedded block.
- FIG. 5 This configuration is illustrated in FIG. 5 which has the same three blocks and scan chain segments as in FIG. 4 and uses the same reference numerals used in the latter to designate like parts. Referring to FIG. 5, it will be seen that internal segments 74 and 80 have their serial inputs connected to the output of segments in an embedded block.
- the source is the output of a peripheral segment, namely segments 76 and 90 , respectively. All other internal blocks are connected to other segments in their respective block.
- the circuit is in IDDq mode when control signals C 1 and C 2 are both active (logic 1).
- the scan chains which are formed in this mode are indicated by the solid lines which interconnect the various segments. The dotted lines indicate other scan test modes.
- peripheral segments 76 and 90 are connected directly to their respective block serial input and are the first segment in their respective scan chains.
- the circuit of FIG. 5 has four chain chains which together collect all memory elements in the circuit.
- a first scan path comprises segment 72 in block 60 ; a second scan path comprises segments 76 , 78 , and 74 ; a third scan path comprises segments 84 , 86 , 90 and 80 ; and a fourth scan path comprises segments 90 , 84 , 86 and 88 .
- control signal C 1 and C 2 define three scan test modes which are used to test the three blocks individually.
- C 1 and C 2 are set to 0,0 to perform a scan test of top level block 60 .
- the scan path formed collects segments 72 , 78 and 74 .
- C 1 and C 2 are set to 0,0 to perform a scan test of block 52 .
- the scan path formed collects segments 76 , 78 , 90 and 80 .
- C 1 and C 2 are set to [0,1] to perform a scan test of embedded block 64 .
- the scan path formed collects segments 84 , 86 , 90 and 88 .
- the scan chains for blocks 60 and 62 include the peripheral segment of their embedded blocks. It will be understood by those skilled in the art that other scan chains may be formed for other tests.
- each block including the top level block, is provided with a single scan chain configuration which collects all scannable memory elements which reside in its block and the serial input of each scan chain is connected to the circuit serial input, the Test Data Input (TDI) of the TAP.
- TTI Test Data Input
- This scan chain includes both internal and peripheral memory elements. All of the scan chains are concurrently enabled in response to a quiescent current test mode signal so that quiescent current test patterns can be concurrently loaded into all blocks in the circuit. Preferably, the same test patterns are concurrently loaded into all blocks and the patterns are pseudo random patterns. However, some blocks may require block specific patterns, as mentioned above.
- the serial input of such a block is connected to a separate source via a circuit primary input. The test patterns from the separate source are loaded into the block concurrently with the loading of test patterns into the other blocks.
- a circuit constructed as described above may have quiescent testing performed as described below with reference to FIG. 6 which illustrates a flow chart of the various steps of the method.
- the circuit is placed in quiescent current test mode (step 100 ) by applying an appropriate instruction to the chip.
- the test controller generates an appropriate Mode signal whereby the serial input of each block is connected to its predetermined circuit primary serial input, such as the TDI pin of the TAP, and the serial input of each block scan chain is connected to the block serial input.
- the serial input of blocks embedded in other blocks i.e. blocks located at one level down in design hierarchy
- This establishes a scan path to all scannable memory elements in all blocks in the circuit and, more specifically, a plurality of parallel scan paths.
- Quiescent current test patterns are then serially loaded through the circuit serial input(s) and into each scan chain in parallel until a predetermined reference point of the test pattern has been reached. This is achieved by setting SE high and activating the clock signal applied to the memory element clock inputs (step 102 ).
- the test patterns will include an initialization section which serve to initialize all memory elements to a known value (step 104 ).
- quiescent current measurement procedures indicated generally by reference numeral 106 , begins.
- Procedure 106 comprises performing a predetermined number of quiescent current measurement steps.
- steps 108 include measuring the quiescent current of the circuit (step 108 ), and storing the measured quiescent current value, and the test pattern point or number at which the value was measured (step 110 ).
- steps 112 As long as the predetermined number of measurements have not been made (step 112 ), an additional sequence of bits of the test pattern is loaded (step 114 ) and the current measurement and storage steps 108 , and 110 , are repeated.
- the procedure is terminated (step 116 ) and the results are analyzed using known methods to determine faults, if any, in the chip.
- the test pattern is a pseudo-random sequence of bits.
- the present invention also contemplates simultaneously loading different sequences of bits, from different sources, into different blocks as well as blocks have more than one quiescent current serial input for loading different sequences of bits into different scan chains of the same block.
- a scan enable signal and a clock signal are used to control scan path segments.
- the scan enable signal and clock signal can be applied directly to circuit inputs or to inputs of the TAP.
- boundary scan it might be necessary to scan predetermined values into the boundary scan register (step 119 ) prior to performing the IDDq measurement steps, to place circuit inputs and outputs in a condition that will not cause an elevated quiescent current.
- FIG. 12 which illustrates the various states of a TAP according to the IEEE 1149 . 1 standard, the scan enable signal is active during the Shift_DR state of the TAP and inactive when the TAP is in another state.
- Clock pulses are generated only during the Capture_DR and Shift_DR states of the TAP.
- the predetermined values are loaded into the boundary scan register by transitioning the TAP through the Select_DR, Capture, Shift, Exit1 and Update states. After shifting for an appropriate number of clock cycles, a measurement can be made while still in the shift cycle or, alternatively, by transitioning the TAP to the Exit1, Pause or Exit2 states. Going for the Pause or Exit states cause the scan enable (SE) signal to go low. This has the merit of testing the memory elements in their functional mode of operation.
- step 114 load more bits
- the quiescent current test method of the present invention employs a scan mode of operation by which quiescent current test patterns are shifted into memory elements of the entire circuit.
- the method is such that it is not only repeatable, but also predictable.
- the operations described below are performed on a description of a circuit under development.
- the first step is to arrange the internal scannable memory elements in each hierarchical block into scan chains according to the rules described earlier with reference to FIG. 4 and 5 and to insert means in the circuit for concurrently configuring input peripheral memory elements in internal test mode and output peripheral memory elements in external test mode. These configurations are based on the principle that the test patterns generated for each block will be applied to the serial input of the block and that the block serial input is ultimately connected to a circuit serial input. This arrangement is necessary to determine a “reference point” of a predetermined sequence of bits applied to the serial input(s) when performing quiescent current measurements, as explained more fully later. This step is performed by an appropriate software tool which determines the appropriate scan chains from the circuit description using the above described segment connection rules.
- each selected block is analyzed to determine whether there exist any circuit states which would cause an elevated quiescent current. As shown in the flow chart of FIG. 7, the analysis is performed on individual blocks in isolation. The analysis begins with the blocks at the lowest level of hierarchy and progresses upwardly to the highest level. This is referred to as a bottoms-up analysis.
- the circuit in FIG. 4 and 5 will be used as an example.
- the lowest level is selected first (step 120 ) and then a block is selected (step 122 ) in the selected level.
- the lowest level is level three which has one block, block 64 .
- Block 64 is selected .
- the block is then analyzed (step 124 ) to determine whether it contains circuit states which would result in elevated IDDq current levels. This is followed by the generation of test patterns (step 126 ) and the calculation of fault coverage (step 128 ) which is stored. Finally, the test patterns are simulated or verified (step 130 ). The patterns are fault simulated using any desired fault model such as a “toggle fault model” or a “pseudo stuck-at fault model”, both of which are well known to those skilled in the art. If there are more blocks in the hierarchical level (step 132 ), the next block is selected (step 134 ) and steps 124 to 130 are repeated. In the illustrated example, there are no other blocks in the third level.
- step 138 the next level is selected (step 138 ) and steps 122 to 136 are repeated.
- the second level is processed and block 62 is selected for analysis.
- block 62 is selected for analysis.
- top level is processed and top level block is selected for analysis.
- quiescent current test patterns are generated and applied to the logic within the block and to the logic at the next level down of design hierarchy via the peripheral memory elements of the blocks selected for analysis.
- Embedded blocks located one level down in design hierarchy, if any, are configured in peripheral memory element quiescent current test mode.
- a simplified or shell model of selected embedded hierarchical blocks is generated for use during analysis of its parent block.
- the simplified model is substituted for the complete description of the embedded block. This substitution simplifies the analysis. Simplified models need not be generated for all embedded blocks, but rather may be limited to certain blocks, such as blocks which exceed a predetermined size.
- the simplified model may contain only the peripheral scannable memory elements or, additionally, some peripheral logic, as described and claimed in Applicant's aforementioned application No. 09/626,877. Further particulars of the analysis procedure are described below in the description of variations of the method which are used for different types of blocks. The analysis is performed using the same test patterns which will be used during actual testing of the circuit and will identify test patterns which must be excluded.
- the fault coverage of the block is first calculated by considering only the logic contained within the block itself and the peripheral logic of embedded blocks located at one level down in the design hierarchy.
- the total fault coverage of the block is calculated by determining the weighted sum of the fault coverage of the block and of its embedded blocks.
- the fault coverage block is weighted by the number of gates contained in each block. It will be noted that the fault coverage of an embedded block can itself be the result of a weighted sum because of embedded blocks that are further down in the design hierarchy.
- the simulation or verification of the application of the test patterns to verify their correctness is optional in a design environment in which the connections of the various scan segments are automated. However, it might be desirable to perform such simulation in a circuit simulator (e.g. Verilog or VHDL) with full timing information to ensure that the circuit will operate as predicted.
- a circuit simulator e.g. Verilog or VHDL
- the steps of analysis, test pattern generation and fault simulation are performed on a circuit representation which includes only the boolean function of circuit gates. Timing considerations are not taken into account.
- the simulation consists of scanning in the test patterns and observing that the desired bit values have been loaded in the correct memory elements at the predicted time. The simulation can be performed either by simulating the entire circuit, i.e. all blocks, simultaneously, or by simulating the host circuit and a simplified model of each of its embedded blocks.
- the most efficient way of organizing the scan path segments inside the embedded blocks is to connect all peripheral scan path segments to a block input or to the output of another peripheral scan path segment so that all peripheral memory elements are at the beginning of the scan path segments and the simulation results of a block do not depend on values shifted through internal scan path segments. Verification can also be performed in a fully hierarchical/manner.
- the first variation is based on the premise that test patterns do not need to be carefully chosen and that, in a good circuit, no circuit states will result in an elevated quiescent current.
- This variation allows a block provider to perform all steps necessary to prepare a block for quiescent current testing, according to the present invention, without having to provide detailed information about the internal structure of the block to the user of the block.
- Both variations use predetermined sequences of bits that are not block specific.
- the same sequence of bits is applied to the serial input of all blocks of these types.
- the sequence is preferably a pseudo-random sequence, but it can be any sequence as long as it is extendable, both forward and backward, from a “reference point”, so that all of the memory elements of the longest chain are initialized.
- a pseudo-random sequence is preferred because of the ease with which it can be to generated algorithmically either on-chip or off-chip.
- Test pattern generation can be implemented by means of either hardware or software. Predetermined sequences can also be stored in tester memory as well.
- the analysis phase determines whether certain circuit states result in elevated quiescent current.
- the result of the analysis must indicate that there are no such states.
- An indication that there are such states means that there is an error in the design of the block and the error must be rectified.
- the analysis can be a static analysis performed independently of the aforementioned predetermined sequence of bits which will applied to the serial inputs during quiescent current testing of the circuit.
- a static analysis assumes that all circuit states are possible. While the analysis is more complex than other types of analyses, it is exhaustive in that it can verify that any predetermined sequence can be applied to the circuit.
- the analysis may, alternatively, consist of a simulation of the logic contained in the block using the predetermined sequence of bits applied to the block serial input during actual quiescent current testing. While a simulation analysis is simpler than a static analysis, it is not exhaustive in that the results obtained with a specific sequence are not necessarily applicable to other sequences.
- the quiescent current test pattern for the circuit is designed so that it includes an initialization section and a quiescent current measuring section.
- the initialization section serves to initialize all scannable memory elements in the circuit to a known value.
- a “reference point” in the predetermined sequence is the point at which quiescent current measuring begins. Assuming a single serial input for each block (the case of multiple serial inputs is discussed later) and a maximum chain length L, the L bits of the predetermined sequence preceding the reference point must be known as these bits will determine the contents of the scan chain.
- the test patterns resulting from shifting the predetermined sequence are calculated and the fault coverage is calculated for each of a predetermined number of current measuring points after the reference point of the predetermined sequence.
- the measuring points are the same for all blocks and can be at regular intervals of the predetermined sequence.
- the PRPG seed corresponds to the PRPG value at the time the first current measurement is performed. Subsequent current measurements are performed at predetermined intervals, such as every 100 clock cycles for as many current measurements as are desired.
- the PRPG is rolled back by N states to determine the starting seed value to load in the PRPG and begin generating the sequence for the block.
- N corresponds to the length of the scan chain.
- the first measurement corresponds to the reference point mentioned previously.
- the first memory element of all chains contains a value corresponding to TDI timeslot 0.
- the next memory element contains TDI timeslot ⁇ 1 and so forth. The same predicted values apply to subsequent measurements and is independent on the length of each scan chain.
- the block with the longest chain will determine the length of the initialization section, i.e., the number of bits which must be provided before the reference point to completely initialize all chains. For example, assume that block 14 has a scan chain length of 30,000 and block 16 has a chain length of 50,000. Block 14 is analyzed using only the 30,000 bits before the reference point to initialize all memory elements. Block 16 is analyzed using the 50,000 bits before the reference point. When the circuit is actually tested, the 50,000 bits before the reference point are applied to the circuit to initialize the longest chain. However, the first 20,000 bits will be ignored by block 14 as it only needs the last 30,000 bits. The last 30,000 bits will correspond exactly to those used to perform the analysis of block 14 in isolation and the 50,000 bits will correspond exactly to those used to perform the analysis of block 16 in isolation.
- the analysis of the blocks is bottom-up. That is, the blocks at the lowest level of the design hierarchy are analyzed first and the information necessary to analyze the block at the next higher level of hierarchy is generated and stored.
- the information would include, for example, the values output by the output peripheral memory elements.
- blocks will have only one block serial input. However, more than one block serial input may be provided if desired without departing from the invention.
- the advantage of using several serial inputs is that it might reduce the test time because it takes less time to load the scan paths. However, the dominant factor is usually the time required to perform the current measurement as opposed to the time required to load the patterns.
- Multiple serial inputs would only be required when the scan chain length of a block is very large (e.g. more than a million memory elements). Individual scan chains in the block would be connected to only one of the block serial inputs.
- the same predetermined sequence of bits can be applied to all inputs or, alternatively, different sequences can be applied to different inputs.
- the use of different sequences for different serial inputs might provide a slightly higher fault coverage.
- the use of the same sequence for all serial inputs simplifies the connections. This is because, if each serial input is associated with a specific predetermined sequence, the connections made at the next level of design hierarchy must take this association into account.
- the same predetermined sequences are used for all embedded blocks. That is, serial inputs of different blocks can be connected together. In this variation, the analysis requires only one pass through the various blocks of the circuit.
- FIGS. 8 - 10 illustrate scan chain configurations which are obtained in accordance with the present invention.
- the three figures illustrate different scan chain configurations for the same circuit.
- Scan chain segments are indicated by an alphabetical character inside a circle. Segments positioned adjacent the perimeter of a block are peripheral memory elements. The arrows indicate scan path shift direction.
- Each figure shows the circuit 10 as having a Test Data Input pin, TDI, which serves as the circuit as well as the block serial input for quiescent current test patterns.
- TDI Test Data Input pin
- host block 12 comprises segments labeled A, B, K, L, M, N, Q, V.
- Block 14 consists of segments C, D, E, F, G, H, I and J.
- Block 16 comprises segments O, P, Q, R, S, T and U.
- a single scan chain configuration is shown for each block. The first segment of each single chain is connected to the TDI pin, which is the serial input of the host block.
- the single scan chain of each block collects all memory elements within its associated block, including both the internal and peripheral memory elements. The order of the memory elements in each segment is extracted for each block separately and stored in a respective block scan chain file in memory.
- FIG. 9 illustrates a single chain configuration which is used when a host or parent block 12 is selected for analysis.
- the single scan chain of the top level block collects all top level memory elements and the peripheral memory elements of collared or embedded hierarchical blocks 14 and 16 .
- FIG. 10 illustrates the scan configuration that is obtained when both embedded blocks 14 and 16 and the top level block 12 are placed into their respective single chain configuration. It will be seen that, in this configuration, all memory elements in all blocks have been collected. This is the scan chain configuration which is used when a quiescent current testing is performed on a physical chip and when the entire chip is analyzed at once. It will be noted that certain peripheral memory elements within hierarchical blocks 14 and 16 , drive two destinations. For example, memory element U drives memory elements S and V. This type of scan chain branching is normally not allowed in scan based tests, but does not create any problems in quiescent current testing because the scan chains are only used to control the states of the memory elements, but not to observe them.
- each memory element in the scan chain is associated to a “timeslot” on TDI except that, due to the branching effect, more than one memory element can be associated with any given timeslot.
- the consequence of the branching is that values within memory elements are forced to be the same for all memory elements associated with a unique timeslot on TDI. This avoids loading of independent values in each hierarchical block.
- the random vector method described herein exploits this feature and allows for fault grading each block independently of all other blocks while obtaining a consolidated fault coverage of the entire chip without ever having to load the entire netlist into any analysis tool.
- blocks 14 and 16 are analyzed before host block is analyzed. Being in the same hierarchical level, the order in which blocks 14 and 16 are analyzed is not important.
- the second variation is based on the premise that test patterns do not need to be carefully chosen and that, in a good circuit, most, but not all, circuit states will not result in elevated quiescent current. Thus, this variation assumes that there may be some states which result in an elevated quiescent current. In contrast with the first variation, the second variation is most useful when the circuit designer has access to detailed information respecting all hierarchical blocks because a fault simulation must be performed on the hierarchical blocks after the blocks have been assembled on the circuit. The method identifies test patterns which result in elevated quiescent current states and excludes them from use during analysis and quiescent current testing of the circuit.
- analysis of the circuit is performed in two passes. Each pass are performed bottom-up, i.e. from the lowest level of design hierarchy to the highest level.
- the first pass identifies the points of the sequence where current measurements can be performed.
- the second pass calculates fault coverage based on the portion of the sequence that were actually used.
- each block is analyzed and a set of constraints is appended to information passed to the next level of design hierarchy.
- the constraints identify the points of the sequence which correspond to elevated quiescent current and also specify test pattern information.
- the constraint information can be appended to a log file produced by an analysis tool.
- the analysis at the next higher level considers the constraints of all embedded blocks and excludes all points of the sequence where at least one block indicated that it corresponded to an elevated quiescent current. This new set of constraints is stored for use in the analysis in the next higher level of design hierarchy where additional points may be excluded based on the analysis at that level.
- test patterns corresponding to the usable points of the sequence are fault simulated and the fault coverage is calculated.
- the blocks at the lowest level of hierarchy are analyzed first and the fault coverage results are stored for use in the next higher level of design hierarchy along with the number of gates in the block so that the total fault coverage of the next higher level can be calculated.
- connections indicated in the description in the claims are “logical” connections meaning that the connections are made during the test mode of interest (i.e. quiescent current measurements). This means that the connections could be made through logic (e.g. multiplexers or other gates) as opposed to direct connections.
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Tests Of Electronic Circuits (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
- Testing Of Individual Semiconductor Devices (AREA)
Abstract
Description
- The present invention relates, in general to the design and testing of integrated circuits and, more specifically, to a method and a program product for designing circuits in a manner to facilitate quiescent current (IDDq) testing of the circuit and to a circuit designed according to the method.
- The complexity of semiconductor circuits often requires partitioning the circuit design into several core modules or blocks, generally referred to as design hierarchy. The blocks can be verified and laid out independently from each other. This design approach has been used for a long time and has several benefits, the most significant of which are to accelerate the development of the design by allowing several designers to work in parallel and to reduce the difficulty for design automation tools by avoiding the need to manipulate the description of the entire circuit at once.
- It has been shown recently that scan test methods can be adapted to a hierarchical design methodology by adding scannable memory elements around the periphery of core modules or by modifying functional memory elements near the periphery of modules and providing an internal and external mode of operation to these memory elements (see, for example, “A structured and Scalable Mechanism for Test Access to Embedded Reusable Cores”, ITC '98.
paper 12. 1, Marinessen et al. and Applicants' co-pending U.S. patent application Ser. No. 09/626,877 filed on Jul. 27, 2000 for “Hierarchical Design and Test Method and System, Program Product Embodying the Method and Integrated Circuit Produced Thereby”, both incorporated herein by reference.) In this divide-and-conquer approach, scan tests of modules can be prepared in advance by the designer of the core module or block without any knowledge of the context within which it will be used. - The preparation of the scan tests involves at least the verification of design-for-test rules and calculation of test coverage. During the internal scan test of blocks, the state of memory elements outside of the modules is irrelevant. The preparation also involves the generation of a simplified model of the module that can be used to perform the same steps at the next level of hierarchy of the design. The use of the simplified model eliminates the need to manipulate the entire description of core modules. During the external scan test of modules, the state of memory elements that are not involved (i.e. not near the periphery of the module) is, again, irrelevant. Core modules can be tested in parallel or sequentially.
- There is one type of test that does not readily lend itself to this hierarchical design methodology. During circuit manufacturing, it is often required to measure the quiescent current (IDDq) for different states of the circuit. The quiescent current is used as a complement to the scan tests described earlier. The state of the circuit is defined by the values loaded in the memory elements of the circuit. The designer needs to ensure that each state of the circuit corresponds to a state where no static current is consumed in a good circuit. A software tool (rules checker) is typically used to perform this task. The designer can also calculate the fault coverage of the faults covered by this test using a fault simulator and/or an automatic test pattern generator (ATPG). To date, the quiescent current test, and the analysis required to perform it, requires considering the entire circuit at once which is becoming virtually impractical for large circuits.
- The present invention provides a hierarchical design method for integrated circuits to be tested using quiescent current monitoring and a circuit designed according to the method. The method allows the state of the entire circuit to be set in a repeatable and predictable manner which avoids the need to analyze the entire circuit at once. The circuit is divided into hierarchical blocks or modules according to a design hierarchy. The blocks have scannable memory elements which are organized into scan chain segments with at least one segment being a peripheral scan chain segment. The peripheral segment is comprised of input and output peripheral scannable memory elements which are configurable in an internal test mode and an external test mode.
- One aspect of the present invention is broadly defined as a method of designing integrated circuits, having at least one hierarchical block, for quiescent current testing, comprising analyzing each of one or more selected hierarchical blocks independently of other selected blocks in the same or higher levels of hierarchy to identify circuit states of each the selected blocks which could result in elevated quiescent current levels during quiescent current testing of the circuit, the analyzing beginning with blocks at a lowest level of design hierarchy and proceeding in sequence through each level of design hierarchy to a highest level of hierarchy containing a top-level block; and calculating a fault coverage of for each the selected block.
- Another aspect of the present invention is generally defined as a circuit having at least one hierarchical block, the circuit comprising: each block having scannable memory elements arranged into two or more scan chain segments having a serial input and a serial output, the memory elements including peripheral memory elements associated with block inputs and outputs, the peripheral scannable memory elements being arranged in peripheral scan chain segments, input peripheral elements being configurable in internal test mode for testing circuitry within the block and output peripheral memory elements being configurable in external test mode for testing circuitry external of the block; means associated with each segment and responsive to an active quiescent current test signal for connecting the serial input of the segment to a block serial input, or to the serial output of another scan chain segment, or to the serial output of a peripheral scan chain segment of a block located at one level of hierarchy down; and means responsive to the active quiescent current test signal for concurrently configuring input peripheral memory elements of the block in internal test mode and configuring output peripheral memory elements of the block in an external test mode during quiescent current testing of the circuit.
- The layout of the circuit does not need to be hierarchical in order to apply the proposed hierarchical quiescent current test method of the present invention. However, if the layout of the circuit is hierarchical, the design hierarchy used for the layout will typically correspond to the top levels of the design hierarchy used for the test method.
- Still another aspect of the present invention is defined as a method of performing quiescent current testing on an integrated circuit having at least one hierarchical logic block, each block having a block serial input, scannable memory elements arranged in at least one scan chain having a scan chain serial input, the method comprising connecting each block serial input to a circuit serial input and connecting each scan chain serial input to its associated block serial input; serially loading a quiescent current test pattern through the circuit serial input so as to load the test pattern into each scan chain in parallel until a predetermined reference point of the test pattern has been reached at which all scannable memory elements in the circuit have been initialized with a known value, and then performing a predetermined number of quiescent current measurement steps including: measuring the quiescent current of the circuit, storing the measured quiescent current value and the test pattern point at which the value was measured; and, while the predetermined number of measurements has not been reached, scanning in an additional sequence of bits of the test pattern and repeating the measurement steps.
- These and other features of the invention will become more apparent from the following description in which reference is made to the appended drawings in which:
- FIG. 1 is a diagrammatic block view of a circuit according to one embodiment of the present invention;
- FIG. 2 is a diagrammatic view of a hierarchical block showing scannable memory elements arranged in scan chains and peripheral memory elements associated with block inputs and outputs;
- FIG. 3 diagrammatically illustrates a prior art scannable memory element;
- FIG. 4 diagrammatically illustrates a circuit having three levels of hierarchy and showing the manner in which scan chain segments may be connected according to one embodiment of the present invention;
- FIG. 5 is similar to FIG. 4, but shows a more restrictive manner in which scan chain segments may be connected according to a preferred embodiment of the present invention;
- FIG. 6 is a flow chart illustrating the method of performing a quiescent current test on a circuit designed according to an embodiment of the method of design of the present invention;
- FIG. 7 is a flow chart generally illustrating a method of designing a hierarchical circuit for quiescent current testing according to an embodiment of the present invention;
- FIG. 8 is a diagram illustrating a single scan chain configuration for two hierarchical blocks located one level down in design hierarchy from a parent or host block;
- FIG. 9 is a diagram similar to FIG. 8, but illustrating a parent block scan chain configuration which includes all scannable memory elements of the parent or top-level block and the peripheral memory elements of two hierarchical block located one level down in design hierarchy;
- FIG. 10 is a diagram similar to FIG. 8, but illustrating a single scan chain configuration for each of a host block and two hierarchical blocks located one level down in design hierarchy, which is used when performing a quiescent current test of the circuit;
- FIG. 11 diagrammatically illustrates the alignment of the serial input of three scan chain segments receiving the same quiescent current test pattern from a pseudo-random pattern generator; and
- FIG. 12 is a state diagram of a Test Access Port according to the IEEE 1149.1 standard.
- In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances, well known methods, procedures, components and circuits have not been described in detail so as not to obscure aspects of the present invention.
- As is known in the art, integrated circuit devices are typically designed and fabricated using one or more computer data files that define the layout of the circuit arrangements of the devices. The files are typically generated by a design tool and are subsequently used during manufacturing to create the layout masks that define the circuit arrangements applied to a semiconductor wafer. Typically, the programs are provided in a predefined format using a hardware description language (HDL) such as VHDL, verilog, EDIF, etc. While the invention has and hereinafter will be described in the context of fully functioning integrated circuit devices and data processing systems utilizing such devices, those skilled in the art will appreciate that the various embodiments of the invention are capable of being distributed as a program product in a variety of forms, and that the invention applies equally regardless of the particular type of signal bearing media used to actually carry out the distribution. Examples of signal bearing media include but are not limited to recordable type media such as volatile and non-volatile memory devices, floppy disks, hard disk drives, CD-ROM's, and DVD's, among others and transmission type media such as digital and analog communications links.
- FIG. 1 illustrates a simple
hierarchical circuit 10 according to one embodiment of the present invention. The circuit includes a top-levelhierarchical block 12, two embedded hierarchicalfunctional blocks boundary scan register 18, a Test Access Port (TAP) 20, and atest controller 22.Reference numeral 24 represents logic in top-level block 12.Blocks blocks - FIG. 2 is a diagrammatic view of the internal structure of a typical hierarchical block. The block generally includes
combinational logic 30 andblock inputs 32 andblock outputs 34, by which the block communicates with other blocks during normal operation of the circuit, and memory orstorage elements 36 which store data during normal operation of the circuit as well as during various test modes of operation. The memory elements are designed to be scannable to allow a test stimulus to be scanned into the elements and a test response data to be scanned out. FIG. 3 illustrates one arrangement of a typicalscannable memory element 36. The element receives input fromcombinational logic 30. The memory element includes a flip-flop 38 having a data input, D, a data output Q and a clock input, CK. Input D is connected to the output of amultiplexer 40. Oneinput 42 of the multiplexer is connected to a memory element serial input, SI. The other input 44 of the multiplexer is connected to an output ofcombinational logic 30. Acontrol input 46 of the multiplexer receives a scan enable signal, SE, which operates to configure the memory element in scan or shift mode when SE is active (logic 1) and in capture or normal mode when SE is inactive (logic 0). Output Q also serves as the memory element Serial Output, SO. - Referring again to FIG. 2, when scan enable SE is active, the scannable memory elements are serially interconnected into one or
more scan chains 48 by connecting the serial output SO of each memory element to the serial input SI of another memory element. The serial input of the first memory element in the chain is connected to a scan chain serial input. The serial output of the last memory element in a scan chain is connected to a scan chain serial output. A scan chain segment is comprised of one or more memory elements. - Memory elements located at the periphery or boundary of a logic block are referred herein to as “peripheral memory elements”, but also known in the art as “partitioning scan cells”. Elements which output a signal from the block, such as
memory elements 50, are output peripheral memory elements. Elements which receive a signal applied to the block, such asmemory elements 52, are input peripheral memory elements. The preferred form of peripheral memory elements are of the type disclosed and claimed in Applicant's above mentioned co-pending U.S. application patent application Ser. No. 09/626,877 which uses functional memory elements to partition a block into an internal partition and an external partition and thereby provide module or block isolation. However, the present invention can be employed in a circuit which uses conventional test dedicated peripheral memory elements. Both forms of peripheral elements are described and illustrated in the aforementioned application. - Output peripheral memory elements are configurable in “external test mode” in which data contained in the memory element is applied to its associated block output node which, in turn, is connected to external circuitry. When the element is not in external test mode, data generated by the block is passed through to the output node. This allows test data to be scanned in to the output peripheral memory elements to test the response of external circuitry to various values of block outputs. Similarly, input peripheral memory elements are configurable in “internal test mode” in which data contained in the peripheral memory element is passed on to the internal circuitry of the block. When the element is not in internal test mode, data applied to its associated block input node is applied to the internal circuitry of the block. This allows test data to be scanned in to the input peripheral memory elements to test the internal logic of the block.
- The top-level block need not necessarily be provided with peripheral memory elements. In this case, the boundary scan register chain can be used as a peripheral chain. However, this is not necessary because access to chip inputs and outputs is available. Peripheral memory elements may be included in the top-level block if the complete circuit may itself constitute a block which is embedded into another circuit.
- Input and output peripheral memory elements are normally not active concurrently. In the present invention, however, the test controller is provided with circuitry (not shown) which responds to an active quiescent current test signal by applying a signal to input and output peripheral memory elements which concurrently configures input peripheral memory elements in internal test mode and output peripheral memory elements in external test mode. This configuration is referred to herein as a “peripheral memory element quiescent current test mode”. This mode allows the peripheral memory elements to be arranged in a scan chain segment and a simplified model of embedded blocks to be substituted for the complete description of the block during analysis of a parent block, as explained more fully later in the description of the method of the present invention. In addition, it allows test data scan in to input and output peripheral memory elements to be concurrently applied to block internal logic as well as to logic external to a block.
- One aspect of the present invention relates to the manner in which internal and peripheral memory elements are arranged into scan chains for use in loading quiescent current test patterns into the various hierarchical blocks in the circuit. In general, memory element segments of a block are arranged, in IDDq test mode, on a scan path which originates from or includes a block serial input. This is achieved by connecting segment serial input to: A) a block input, B) the serial output of another segment or C) the serial output of a segment located in an embedded block one level down in design hierarchy and by connecting the block serial input of embedded blocks to the block serial input A segment comprises one or more serially configurable scannable memory elements.
- This configuration allows the use of a simplified or shell model of a block during a block analysis design phase, described later. The simplified model contains peripheral memory element segments and peripheral logic and additional files which describe the scan paths of the block and the scan paths of embedded blocks and allows for pattern generation. In this arrangement, some rules checking (e.g. scan path continuity) and verification/simulation is performed in an indirect manner in which other scan test modes are rules checked and verified/simulated and it assumed that the IDDq test mode is correct by construction. Thus, this general configuration does not place any restrictions on internal and peripheral segments.
- This configuration also allows the use of simplified models which contain all scan paths and peripheral logic which, in turn, allows for a complete analysis including rules checking, pattern generation and verification (i.e. simulation). The configuration provides significant size reduction in the size of simplified models, but is not optimal.
- This general configuration will be better understood by referring to FIG. 4 which illustrates a circuit having three levels of hierarchy. The figure shows three blocks including a
top level block 60 having an embeddedblock 62 which defines a second level of hierarchy and which, in turn, has an embeddedblock 64 which defines a third level of hierarchy. Scan chain segments are shown as a sequence of rectangles which represent memory elements. Peripheral segments are bolded. A segment may consist of one or more scannable memory elements.Blocks serial inputs Host block 60 includes twointernal segments Segment 72 is connected to top level blockserial input 66, the chip circuit input. Embeddedblock 62 includes twointernal segments peripheral segment 78. In this simple example, block 62 is provided with means, in the form of amultiplexer 82, for selectively connecting the serial input ofperipheral segment 78 to one of two sources including the output ofinternal segment 76 ofblock 62 or to the serial output ofinternal segment 72 ofparent block 60.Multiplexer 82 is controlled by a control signal Cl. The output ofperipheral segment 78 is connected to the input ofinternal segment 74 ofparent block 60. - Embedded
block 64, which resides in the third level of hierarchy, has four segments, includinginternal segments peripheral segment 90.Segments segment 84 being connected to blockserial input 70. Amultiplexer 92 receives the serial output ofperipheral segment 78 inparent block 62 and the output ofinternal segment 86 and applies its output toperipheral segment 90.Multiplexer 92 is controlled by control signal C2. The serial output ofperipheral segment 90 branches tointernal segment 80 ofparent block 62 andinternal segment 88 ofcurrent block 64. It will be seen that all segments reside on scan path which originates from or includes a block serial input. The serial input ofblock 64 is connected to the serial input ofparent block 62. Control signals C1 and C2 are mode signals which are used to define different scan chains configurations for scan testing and quiescent current testing of the circuit. A specific example of various modes is given below with reference to FIG. 5. - It will be seen that the general configuration described above allows segments which depend on internal segments of embedded blocks. In some cases, this can lead to complications with regard to generating and verifying test patterns to be used during quiescent current testing. The situation can be simplified by placing restrictions on the connections of internal and peripheral scan chain segments as described below.
- In IDDq test mode, internal segments are arranged on a scan path which originates from or includes a block serial input by connecting the internal chain segment serial input to: A) a block input, B) the serial output of another segment (internal or peripheral) in the block or C) the serial output of a peripheral segment located in an embedded block one level down in design hierarchy. Thus, the primary difference between this configuration and the general configuration described above is with regard to the option (C). More specifically, if an internal segment is connected to any segment in an embedded block, it must be connected to the output of a peripheral segment.
- In IDDq test mode, peripheral segments are arranged on a scan path which originates from or includes a block serial input by connecting the peripheral segment serial input to: A) a block input, B) the serial output of another peripheral segment of the block or C) the serial output of a peripheral segment located in an embedded block. Thus, this limits the sources for peripheral segments to either the block serial input or to the output of another peripheral segment in its resident block or in an embedded block. This configuration is illustrated in FIG. 5 which has the same three blocks and scan chain segments as in FIG. 4 and uses the same reference numerals used in the latter to designate like parts. Referring to FIG. 5, it will be seen that
internal segments segments peripheral segments segment 72 inblock 60; a second scan path comprisessegments segments segments - In addition to defining an IDDq mode, control signal C1 and C2, define three scan test modes which are used to test the three blocks individually. C1 and C2 are set to 0,0 to perform a scan test of
top level block 60. The scan path formed collectssegments block 52. The scan path formed collectssegments block 64. The scan path formed collectssegments blocks - The specific inter-block connections shown in FIG. 4 and5 are for illustration purposes to show some possible connections. In the preferred form of the invention, each block, including the top level block, is provided with a single scan chain configuration which collects all scannable memory elements which reside in its block and the serial input of each scan chain is connected to the circuit serial input, the Test Data Input (TDI) of the TAP. This is illustrated in FIG. 11. This scan chain includes both internal and peripheral memory elements. All of the scan chains are concurrently enabled in response to a quiescent current test mode signal so that quiescent current test patterns can be concurrently loaded into all blocks in the circuit. Preferably, the same test patterns are concurrently loaded into all blocks and the patterns are pseudo random patterns. However, some blocks may require block specific patterns, as mentioned above. The serial input of such a block is connected to a separate source via a circuit primary input. The test patterns from the separate source are loaded into the block concurrently with the loading of test patterns into the other blocks.
- A circuit constructed as described above may have quiescent testing performed as described below with reference to FIG. 6 which illustrates a flow chart of the various steps of the method. The circuit is placed in quiescent current test mode (step100) by applying an appropriate instruction to the chip. In response, the test controller generates an appropriate Mode signal whereby the serial input of each block is connected to its predetermined circuit primary serial input, such as the TDI pin of the TAP, and the serial input of each block scan chain is connected to the block serial input. In the simplest form of the method, the serial input of blocks embedded in other blocks (i.e. blocks located at one level down in design hierarchy), are connected to the block serial input of their respective parent block. This establishes a scan path to all scannable memory elements in all blocks in the circuit and, more specifically, a plurality of parallel scan paths.
- Quiescent current test patterns are then serially loaded through the circuit serial input(s) and into each scan chain in parallel until a predetermined reference point of the test pattern has been reached. This is achieved by setting SE high and activating the clock signal applied to the memory element clock inputs (step102). The test patterns will include an initialization section which serve to initialize all memory elements to a known value (step 104). When the sequence of bits of the test pattern reaches a certain predetermined point, referred to herein as a “reference point”, quiescent current measurement procedures, indicated generally by
reference numeral 106, begins.Procedure 106 comprises performing a predetermined number of quiescent current measurement steps. These steps include measuring the quiescent current of the circuit (step 108), and storing the measured quiescent current value, and the test pattern point or number at which the value was measured (step 110). As long as the predetermined number of measurements have not been made (step 112), an additional sequence of bits of the test pattern is loaded (step 114) and the current measurement andstorage steps - The present invention also contemplates simultaneously loading different sequences of bits, from different sources, into different blocks as well as blocks have more than one quiescent current serial input for loading different sequences of bits into different scan chains of the same block.
- As mentioned, during testing, a scan enable signal and a clock signal are used to control scan path segments. The scan enable signal and clock signal can be applied directly to circuit inputs or to inputs of the TAP. When a TAP is provided and boundary scan is used in the top-level block, it might be necessary to scan predetermined values into the boundary scan register (step119) prior to performing the IDDq measurement steps, to place circuit inputs and outputs in a condition that will not cause an elevated quiescent current. Referring to FIG. 12, which illustrates the various states of a TAP according to the IEEE 1149.1 standard, the scan enable signal is active during the Shift_DR state of the TAP and inactive when the TAP is in another state. Clock pulses are generated only during the Capture_DR and Shift_DR states of the TAP. The predetermined values are loaded into the boundary scan register by transitioning the TAP through the Select_DR, Capture, Shift, Exit1 and Update states. After shifting for an appropriate number of clock cycles, a measurement can be made while still in the shift cycle or, alternatively, by transitioning the TAP to the Exit1, Pause or Exit2 states. Going for the Pause or Exit states cause the scan enable (SE) signal to go low. This has the merit of testing the memory elements in their functional mode of operation. After a measurement has been made, step 114 (load more bits) is performed by transitioning back to the Shift state without traversing the Update state. This preserves the predetermined values loaded in
step 119. In this way, once a test pattern has been shifted in, the current measurements are taken for various combinations of the scan enable and clock signals to maximize the fault coverage. This can also be achieved in a circuit which does not have a TAP. - The description now turns to the method of designing blocks to enable a circuit for quiescent current testing according to the present invention. As is evident from the foregoing, the quiescent current test method of the present invention employs a scan mode of operation by which quiescent current test patterns are shifted into memory elements of the entire circuit. The method is such that it is not only repeatable, but also predictable. The operations described below are performed on a description of a circuit under development.
- It is assumed at the outset that functional memory elements in all blocks have been converted to scannable memory elements to include a multiplexer, such as
multiplexer 40 in FIG. 3 so that the memory elements are configurable in scan mode and capture mode, and provision has been made to provide peripheral memory elements. Test dedicated memory elements may be inserted between the output of combinational logic of the block and the output node of the block. Alternatively, as described and claimed in Applicant's co-pending application, supra, functional memory elements at the periphery or boundary of each block are converted to peripheral memory elements. In the illustrated examples of FIG. 1, 4 and 5, each of the blocks shown therein have been so modified. - The following steps are performed on selected blocks in the circuit. It will be understood by those skilled in the art that any sub-circuit may be selected or designated as a hierarchical block. Each selected block is provided with peripheral memory elements.
- The first step is to arrange the internal scannable memory elements in each hierarchical block into scan chains according to the rules described earlier with reference to FIG. 4 and5 and to insert means in the circuit for concurrently configuring input peripheral memory elements in internal test mode and output peripheral memory elements in external test mode. These configurations are based on the principle that the test patterns generated for each block will be applied to the serial input of the block and that the block serial input is ultimately connected to a circuit serial input. This arrangement is necessary to determine a “reference point” of a predetermined sequence of bits applied to the serial input(s) when performing quiescent current measurements, as explained more fully later. This step is performed by an appropriate software tool which determines the appropriate scan chains from the circuit description using the above described segment connection rules.
- Once the scan chain segments and chains in a hierarchical block have been defined and the block descriptions modified, each selected block is analyzed to determine whether there exist any circuit states which would cause an elevated quiescent current. As shown in the flow chart of FIG. 7, the analysis is performed on individual blocks in isolation. The analysis begins with the blocks at the lowest level of hierarchy and progresses upwardly to the highest level. This is referred to as a bottoms-up analysis. The circuit in FIG. 4 and5 will be used as an example. Thus, the lowest level is selected first (step 120) and then a block is selected (step 122) in the selected level. In the example, the lowest level is level three which has one block, block 64.
Block 64 is selected . The block is then analyzed (step 124) to determine whether it contains circuit states which would result in elevated IDDq current levels. This is followed by the generation of test patterns (step 126) and the calculation of fault coverage (step 128) which is stored. Finally, the test patterns are simulated or verified (step 130). The patterns are fault simulated using any desired fault model such as a “toggle fault model” or a “pseudo stuck-at fault model”, both of which are well known to those skilled in the art. If there are more blocks in the hierarchical level (step 132), the next block is selected (step 134) and steps 124 to 130 are repeated. In the illustrated example, there are no other blocks in the third level. If all blocks in the current level have been processed and there are more levels to analyze (step 136), the next level is selected (step 138) and steps 122 to 136 are repeated. In the example, the second level is processed and block 62 is selected for analysis. Onceblock 62 has been analyzed, the top level is processed and top level block is selected for analysis. When all levels have been processed, the procedure is complete. The above steps will now be described in more detail. - When analyzing the circuit description (typically, a gate level description) of an individual block during the design process, quiescent current test patterns are generated and applied to the logic within the block and to the logic at the next level down of design hierarchy via the peripheral memory elements of the blocks selected for analysis. Embedded blocks located one level down in design hierarchy, if any, are configured in peripheral memory element quiescent current test mode.
- Preferably, a simplified or shell model of selected embedded hierarchical blocks is generated for use during analysis of its parent block. The simplified model is substituted for the complete description of the embedded block. This substitution simplifies the analysis. Simplified models need not be generated for all embedded blocks, but rather may be limited to certain blocks, such as blocks which exceed a predetermined size. The simplified model may contain only the peripheral scannable memory elements or, additionally, some peripheral logic, as described and claimed in Applicant's aforementioned application No. 09/626,877. Further particulars of the analysis procedure are described below in the description of variations of the method which are used for different types of blocks. The analysis is performed using the same test patterns which will be used during actual testing of the circuit and will identify test patterns which must be excluded.
- For a block which contains embedded blocks, the fault coverage of the block is first calculated by considering only the logic contained within the block itself and the peripheral logic of embedded blocks located at one level down in the design hierarchy. The total fault coverage of the block is calculated by determining the weighted sum of the fault coverage of the block and of its embedded blocks. The fault coverage block is weighted by the number of gates contained in each block. It will be noted that the fault coverage of an embedded block can itself be the result of a weighted sum because of embedded blocks that are further down in the design hierarchy.
- The simulation or verification of the application of the test patterns to verify their correctness, is optional in a design environment in which the connections of the various scan segments are automated. However, it might be desirable to perform such simulation in a circuit simulator (e.g. Verilog or VHDL) with full timing information to ensure that the circuit will operate as predicted.
- The steps of analysis, test pattern generation and fault simulation are performed on a circuit representation which includes only the boolean function of circuit gates. Timing considerations are not taken into account. The simulation consists of scanning in the test patterns and observing that the desired bit values have been loaded in the correct memory elements at the predicted time. The simulation can be performed either by simulating the entire circuit, i.e. all blocks, simultaneously, or by simulating the host circuit and a simplified model of each of its embedded blocks. When a simplified model is used for embedded blocks, the most efficient way of organizing the scan path segments inside the embedded blocks is to connect all peripheral scan path segments to a block input or to the output of another peripheral scan path segment so that all peripheral memory elements are at the beginning of the scan path segments and the simulation results of a block do not depend on values shifted through internal scan path segments. Verification can also be performed in a fully hierarchical/manner.
- As mentioned above, certain steps of the methods of designing differ slightly from one another, depending on the nature or type of block. The following description addresses these variations. There are two variations of the basic method that will be described. Both variations rely on the application of a standard predetermined sequence of bits to the serial inputs of blocks. Preferably, all blocks have a single serial input so that all blocks can be connected to the same test pattern source. The first variation is based on the premise that test patterns do not need to be carefully chosen and that, in a good circuit, no circuit states result in elevated quiescent current. The second variation is based on the premise that test patterns do not need to be carefully chosen and that, in a good circuit, most circuit states will not result in elevated quiescent current. Thus, this variation contemplates that some states result in an elevated quiescent current and includes steps to identify test patterns which create elevated quiescent current conditions.
- First Variation
- As previously mentioned, the first variation is based on the premise that test patterns do not need to be carefully chosen and that, in a good circuit, no circuit states will result in an elevated quiescent current. This variation allows a block provider to perform all steps necessary to prepare a block for quiescent current testing, according to the present invention, without having to provide detailed information about the internal structure of the block to the user of the block.
- Both variations use predetermined sequences of bits that are not block specific. The same sequence of bits is applied to the serial input of all blocks of these types. The sequence is preferably a pseudo-random sequence, but it can be any sequence as long as it is extendable, both forward and backward, from a “reference point”, so that all of the memory elements of the longest chain are initialized. A pseudo-random sequence is preferred because of the ease with which it can be to generated algorithmically either on-chip or off-chip. Test pattern generation can be implemented by means of either hardware or software. Predetermined sequences can also be stored in tester memory as well.
- The analysis phase determines whether certain circuit states result in elevated quiescent current. In the first variation of the method, the result of the analysis must indicate that there are no such states. An indication that there are such states means that there is an error in the design of the block and the error must be rectified. The analysis can be a static analysis performed independently of the aforementioned predetermined sequence of bits which will applied to the serial inputs during quiescent current testing of the circuit. A static analysis assumes that all circuit states are possible. While the analysis is more complex than other types of analyses, it is exhaustive in that it can verify that any predetermined sequence can be applied to the circuit. The analysis may, alternatively, consist of a simulation of the logic contained in the block using the predetermined sequence of bits applied to the block serial input during actual quiescent current testing. While a simulation analysis is simpler than a static analysis, it is not exhaustive in that the results obtained with a specific sequence are not necessarily applicable to other sequences.
- The quiescent current test pattern for the circuit is designed so that it includes an initialization section and a quiescent current measuring section. The initialization section serves to initialize all scannable memory elements in the circuit to a known value. A “reference point” in the predetermined sequence is the point at which quiescent current measuring begins. Assuming a single serial input for each block (the case of multiple serial inputs is discussed later) and a maximum chain length L, the L bits of the predetermined sequence preceding the reference point must be known as these bits will determine the contents of the scan chain.
- Starting from the reference point of the predetermined sequence, the test patterns resulting from shifting the predetermined sequence are calculated and the fault coverage is calculated for each of a predetermined number of current measuring points after the reference point of the predetermined sequence. The measuring points are the same for all blocks and can be at regular intervals of the predetermined sequence. By aligning the reference point, i.e. by concurrently loading the same test pattern into all quiescent current scan paths in all blocks, it is possible to perform the analysis of each block in isolation without knowing anything about the other blocks. The procedure to generate the test pattern for each block is performed in such a way that it will be possible to apply the patterns after the block has been inserted on a chip. The procedure is as follows. A Pseudo Random Pattern Generator (PRPG) sequence (see FIG. 11) is applied with a constant, predetermined polynomial and seed to all blocks. The PRPG seed corresponds to the PRPG value at the time the first current measurement is performed. Subsequent current measurements are performed at predetermined intervals, such as every 100 clock cycles for as many current measurements as are desired. Depending on the length of the scan chain of a given block, the PRPG is rolled back by N states to determine the starting seed value to load in the PRPG and begin generating the sequence for the block. N corresponds to the length of the scan chain. This approach of aligning the seed at the first measurement instead of at the beginning of the sequence allows aligning the sequence for multiple blocks which have different scan chain lengths. The longest sequence is chosen so that all blocks end up in the same predicted state at the time of the first (the reference point of the sequence) and the subsequent current measurements. Extra TDI values are simply flushed out at the other end for the smaller scan chains.
- The first measurement corresponds to the reference point mentioned previously. At the time of the first measurement, the first memory element of all chains contains a value corresponding to
TDI timeslot 0. The next memory element contains TDI timeslot −1 and so forth. The same predicted values apply to subsequent measurements and is independent on the length of each scan chain. - When the various blocks are integrated into a circuit, the block with the longest chain will determine the length of the initialization section, i.e., the number of bits which must be provided before the reference point to completely initialize all chains. For example, assume that
block 14 has a scan chain length of 30,000 and block 16 has a chain length of 50,000.Block 14 is analyzed using only the 30,000 bits before the reference point to initialize all memory elements.Block 16 is analyzed using the 50,000 bits before the reference point. When the circuit is actually tested, the 50,000 bits before the reference point are applied to the circuit to initialize the longest chain. However, the first 20,000 bits will be ignored byblock 14 as it only needs the last 30,000 bits. The last 30,000 bits will correspond exactly to those used to perform the analysis ofblock 14 in isolation and the 50,000 bits will correspond exactly to those used to perform the analysis ofblock 16 in isolation. - From the reference point, current measurements are performed every time a certain number of new bits of the sequence have been shifted in. This number is arbitrary. However, a standard number can be used so that the fault coverage can be computed in advance for all blocks. For example, a measurement interval of every 100 bits of the predetermined sequence would be acceptable.
- The analysis of the blocks is bottom-up. That is, the blocks at the lowest level of the design hierarchy are analyzed first and the information necessary to analyze the block at the next higher level of hierarchy is generated and stored. The information would include, for example, the values output by the output peripheral memory elements.
- Generally, blocks will have only one block serial input. However, more than one block serial input may be provided if desired without departing from the invention. The advantage of using several serial inputs is that it might reduce the test time because it takes less time to load the scan paths. However, the dominant factor is usually the time required to perform the current measurement as opposed to the time required to load the patterns. Multiple serial inputs would only be required when the scan chain length of a block is very large (e.g. more than a million memory elements). Individual scan chains in the block would be connected to only one of the block serial inputs.
- The same predetermined sequence of bits can be applied to all inputs or, alternatively, different sequences can be applied to different inputs. The use of different sequences for different serial inputs might provide a slightly higher fault coverage. The use of the same sequence for all serial inputs simplifies the connections. This is because, if each serial input is associated with a specific predetermined sequence, the connections made at the next level of design hierarchy must take this association into account. The same predetermined sequences are used for all embedded blocks. That is, serial inputs of different blocks can be connected together. In this variation, the analysis requires only one pass through the various blocks of the circuit.
- FIGS.8-10 illustrate scan chain configurations which are obtained in accordance with the present invention. The three figures illustrate different scan chain configurations for the same circuit. Scan chain segments are indicated by an alphabetical character inside a circle. Segments positioned adjacent the perimeter of a block are peripheral memory elements. The arrows indicate scan path shift direction. Each figure shows the
circuit 10 as having a Test Data Input pin, TDI, which serves as the circuit as well as the block serial input for quiescent current test patterns. - Referring to FIG. 8,
host block 12 comprises segments labeled A, B, K, L, M, N, Q,V. Block 14 consists of segments C, D, E, F, G, H, I andJ. Block 16 comprises segments O, P, Q, R, S, T and U. A single scan chain configuration is shown for each block. The first segment of each single chain is connected to the TDI pin, which is the serial input of the host block. The single scan chain of each block collects all memory elements within its associated block, including both the internal and peripheral memory elements. The order of the memory elements in each segment is extracted for each block separately and stored in a respective block scan chain file in memory. - FIG. 9 illustrates a single chain configuration which is used when a host or
parent block 12 is selected for analysis. The single scan chain of the top level block collects all top level memory elements and the peripheral memory elements of collared or embeddedhierarchical blocks - FIG. 10 illustrates the scan configuration that is obtained when both embedded
blocks top level block 12 are placed into their respective single chain configuration. It will be seen that, in this configuration, all memory elements in all blocks have been collected. This is the scan chain configuration which is used when a quiescent current testing is performed on a physical chip and when the entire chip is analyzed at once. It will be noted that certain peripheral memory elements withinhierarchical blocks block 16 and can be made easily available during the analysis ofparent block 12. - As with a normal scan chain, the position of each memory element in the scan chain is associated to a “timeslot” on TDI except that, due to the branching effect, more than one memory element can be associated with any given timeslot. The consequence of the branching is that values within memory elements are forced to be the same for all memory elements associated with a unique timeslot on TDI. This avoids loading of independent values in each hierarchical block. The random vector method described herein exploits this feature and allows for fault grading each block independently of all other blocks while obtaining a consolidated fault coverage of the entire chip without ever having to load the entire netlist into any analysis tool.
- When performing the analysis of this circuit, blocks14 and 16 are analyzed before host block is analyzed. Being in the same hierarchical level, the order in which blocks 14 and 16 are analyzed is not important.
- Second Variation
- The second variation is based on the premise that test patterns do not need to be carefully chosen and that, in a good circuit, most, but not all, circuit states will not result in elevated quiescent current. Thus, this variation assumes that there may be some states which result in an elevated quiescent current. In contrast with the first variation, the second variation is most useful when the circuit designer has access to detailed information respecting all hierarchical blocks because a fault simulation must be performed on the hierarchical blocks after the blocks have been assembled on the circuit. The method identifies test patterns which result in elevated quiescent current states and excludes them from use during analysis and quiescent current testing of the circuit.
- In this variation, analysis of the circuit is performed in two passes. Each pass are performed bottom-up, i.e. from the lowest level of design hierarchy to the highest level. The first pass identifies the points of the sequence where current measurements can be performed. The second pass calculates fault coverage based on the portion of the sequence that were actually used.
- During the first pass, starting with the blocks at the lowest level of design hierarchy, each block is analyzed and a set of constraints is appended to information passed to the next level of design hierarchy. The constraints identify the points of the sequence which correspond to elevated quiescent current and also specify test pattern information. The constraint information can be appended to a log file produced by an analysis tool. The analysis at the next higher level considers the constraints of all embedded blocks and excludes all points of the sequence where at least one block indicated that it corresponded to an elevated quiescent current. This new set of constraints is stored for use in the analysis in the next higher level of design hierarchy where additional points may be excluded based on the analysis at that level. Once the top and final level block has been analyzed, all usable, or non-usable, points of the sequence have been identified.
- During the second pass, test patterns corresponding to the usable points of the sequence are fault simulated and the fault coverage is calculated. Again, the blocks at the lowest level of hierarchy are analyzed first and the fault coverage results are stored for use in the next higher level of design hierarchy along with the number of gates in the block so that the total fault coverage of the next higher level can be calculated.
- All the connections indicated in the description in the claims are “logical” connections meaning that the connections are made during the test mode of interest (i.e. quiescent current measurements). This means that the connections could be made through logic (e.g. multiplexers or other gates) as opposed to direct connections.
- Although the present invention has been described in detail with regard to preferred embodiments and drawings of the invention, it will be apparent to those skilled in the art that various adaptions, modifications and alterations may be accomplished without departing from the spirit and scope of the present invention. Accordingly, it is to be understood that the accompanying drawings as set forth herein above are not intended to limit the breadth of the present invention, which should be inferred only from the following claims and their appropriately construed legal equivalents.
Claims (82)
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CA002360291A CA2360291A1 (en) | 2001-10-30 | 2001-10-30 | Method and program product for designing hierarchical circuit for quiescent current testing and circuit produced thereby |
US10/011,128 US6868532B2 (en) | 2001-10-30 | 2001-12-10 | Method and program product for designing hierarchical circuit for quiescent current testing and circuit produced thereby |
PCT/CA2002/001582 WO2003038450A2 (en) | 2001-10-30 | 2002-10-22 | Method and program product for designing hierarchical circuit for quiescent current testing |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CA002360291A CA2360291A1 (en) | 2001-10-30 | 2001-10-30 | Method and program product for designing hierarchical circuit for quiescent current testing and circuit produced thereby |
US10/011,128 US6868532B2 (en) | 2001-10-30 | 2001-12-10 | Method and program product for designing hierarchical circuit for quiescent current testing and circuit produced thereby |
Publications (2)
Publication Number | Publication Date |
---|---|
US20030110457A1 true US20030110457A1 (en) | 2003-06-12 |
US6868532B2 US6868532B2 (en) | 2005-03-15 |
Family
ID=27805899
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/011,128 Expired - Lifetime US6868532B2 (en) | 2001-10-30 | 2001-12-10 | Method and program product for designing hierarchical circuit for quiescent current testing and circuit produced thereby |
Country Status (3)
Country | Link |
---|---|
US (1) | US6868532B2 (en) |
CA (1) | CA2360291A1 (en) |
WO (1) | WO2003038450A2 (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030069724A1 (en) * | 1999-11-30 | 2003-04-10 | Bridges2Silicon, Inc. | Method and system for debugging an electronic system using instrumentation circuitry and a logic analyzer |
US20030115522A1 (en) * | 2001-12-17 | 2003-06-19 | Benoit Nadeau-Dostie | Method and program product for designing hierarchical circuit for quiescent current testing |
US20030131325A1 (en) * | 1999-11-30 | 2003-07-10 | Bridges2Silicon, Inc. | Method and user interface for debugging an electronic system |
US6769101B2 (en) * | 2002-05-13 | 2004-07-27 | Agilent Technologies, Inc. | Systems and methods providing scan-based delay test generation |
US20040187060A1 (en) * | 2003-03-21 | 2004-09-23 | Rohrbaugh John G. | Generating test patterns for testing an integrated circuit |
US20050010880A1 (en) * | 1999-11-30 | 2005-01-13 | Bridges2Silicon, Inc. | Method and user interface for debugging an electronic system |
US20060259838A1 (en) * | 2005-05-13 | 2006-11-16 | Lewis Nardini | Scan Sequenced Power-On Initialization |
WO2006138488A2 (en) * | 2005-06-15 | 2006-12-28 | Mentor Graphics Corporation | Reduced-pin-count-testing architectures for applying test patterns |
US7308656B1 (en) * | 2005-10-04 | 2007-12-11 | Xilinx, Inc. | Method and apparatus for generating a boundary scan description and model |
US7827510B1 (en) | 2002-06-07 | 2010-11-02 | Synopsys, Inc. | Enhanced hardware debugging with embedded FPGAS in a hardware description language |
WO2011014203A1 (en) * | 2009-07-28 | 2011-02-03 | Synopsys, Inc. | Hierarchical order ranked simulation of electronic circuits |
US20110121838A1 (en) * | 2009-11-25 | 2011-05-26 | International Business Machines Corporation | Method and apparatus for increased effectiveness of delay and transistion fault testing |
US20110314348A1 (en) * | 2006-02-03 | 2011-12-22 | Texas Instruments Incorporated | Scan testing using scan frames with embedded commands |
US20150043291A1 (en) * | 2013-08-09 | 2015-02-12 | SK Hynix Inc. | Method for testing semiconductor apparatus and test system using the same |
WO2023070301A1 (en) * | 2021-10-26 | 2023-05-04 | 华为技术有限公司 | Method, apparatus and device for logic simulation |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040172382A1 (en) * | 2003-02-28 | 2004-09-02 | Microsoft Corporation. | System and method for generating a request for information about selected objects |
TW200426574A (en) * | 2003-05-21 | 2004-12-01 | Spirox Corp | Fault pattern oriented defect diagnosis for memories |
US7493578B1 (en) * | 2005-03-18 | 2009-02-17 | Xilinx, Inc. | Correlation of data from design analysis tools with design blocks in a high-level modeling system |
US7496869B1 (en) | 2005-10-04 | 2009-02-24 | Xilinx, Inc. | Method and apparatus for implementing a program language description of a circuit design for an integrated circuit |
US7363599B1 (en) | 2005-10-04 | 2008-04-22 | Xilinx, Inc. | Method and system for matching a hierarchical identifier |
US9104894B2 (en) | 2005-12-16 | 2015-08-11 | Hewlett-Packard Development Company, L.P. | Hardware enablement using an interface |
US7380232B1 (en) | 2006-03-10 | 2008-05-27 | Xilinx, Inc. | Method and apparatus for designing a system for implementation in a programmable logic device |
US7761272B1 (en) | 2006-03-10 | 2010-07-20 | Xilinx, Inc. | Method and apparatus for processing a dataflow description of a digital processing system |
US8402409B1 (en) | 2006-03-10 | 2013-03-19 | Xilinx, Inc. | Method and apparatus for supporting run-time reconfiguration in a programmable logic integrated circuit |
US7823103B2 (en) * | 2006-10-24 | 2010-10-26 | International Business Machines Corporation | Method and system of introducing hierarchy into design rule checking test cases and rotation of test case data |
US7673206B2 (en) * | 2007-09-14 | 2010-03-02 | Tilera Corporation | Method and system for routing scan chains in an array of processor resources |
US9057764B2 (en) * | 2011-10-27 | 2015-06-16 | International Business Machines Corporation | Detection of unchecked signals in circuit design verification |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5067091A (en) * | 1988-01-21 | 1991-11-19 | Kabushiki Kaisha Toshiba | Circuit design conversion apparatus |
US5459737A (en) * | 1993-07-07 | 1995-10-17 | National Semiconductor Corporation | Test access port controlled built in current monitor for IC devices |
US5796990A (en) * | 1995-06-02 | 1998-08-18 | International Business Machines Corporation | Hierarchical fault modeling system and method |
US5949692A (en) * | 1996-08-28 | 1999-09-07 | Synopsys, Inc. | Hierarchical scan architecture for design for test applications |
US5987636A (en) * | 1998-05-28 | 1999-11-16 | Nec Usa, Inc. | Static test sequence compaction using two-phase restoration and segment manipulation |
US6151694A (en) * | 1997-11-17 | 2000-11-21 | Kabushiki Kaisha Toshiba | Method of evaluating fault coverage |
US6487688B1 (en) * | 1999-12-23 | 2002-11-26 | Logicvision, Inc. | Method for testing circuits with tri-state drivers and circuit for use therewith |
US6516432B1 (en) * | 1999-12-22 | 2003-02-04 | International Business Machines Corporation | AC scan diagnostic method |
US6718524B1 (en) * | 2001-09-17 | 2004-04-06 | Lsi Logic Corporation | Method and apparatus for estimating state-dependent gate leakage in an integrated circuit |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5592493A (en) | 1994-09-13 | 1997-01-07 | Motorola Inc. | Serial scan chain architecture for a data processing system and method of operation |
US5760598A (en) | 1996-02-12 | 1998-06-02 | International Business Machines Corporation | Method and apparatus for testing quiescent current in integrated circuits |
JP2947204B2 (en) | 1997-02-24 | 1999-09-13 | 日本電気株式会社 | LSI failure location identification method |
US6175244B1 (en) | 1997-04-25 | 2001-01-16 | Carnegie Mellon University | Current signatures for IDDQ testing |
JP3689537B2 (en) | 1997-08-26 | 2005-08-31 | 株式会社リコー | IDDQ test cycle selective extraction device |
KR100567936B1 (en) | 1997-10-31 | 2006-04-07 | 코닌클리케 필립스 일렉트로닉스 엔.브이. | Core test control |
JP3669836B2 (en) | 1998-03-19 | 2005-07-13 | 株式会社リコー | IDDQ test cycle selective extraction device |
PL350155A1 (en) | 1998-09-30 | 2002-11-18 | Cadence Design Systems | Block based design methodology |
-
2001
- 2001-10-30 CA CA002360291A patent/CA2360291A1/en not_active Abandoned
- 2001-12-10 US US10/011,128 patent/US6868532B2/en not_active Expired - Lifetime
-
2002
- 2002-10-22 WO PCT/CA2002/001582 patent/WO2003038450A2/en not_active Application Discontinuation
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5067091A (en) * | 1988-01-21 | 1991-11-19 | Kabushiki Kaisha Toshiba | Circuit design conversion apparatus |
US5459737A (en) * | 1993-07-07 | 1995-10-17 | National Semiconductor Corporation | Test access port controlled built in current monitor for IC devices |
US5796990A (en) * | 1995-06-02 | 1998-08-18 | International Business Machines Corporation | Hierarchical fault modeling system and method |
US5949692A (en) * | 1996-08-28 | 1999-09-07 | Synopsys, Inc. | Hierarchical scan architecture for design for test applications |
US6151694A (en) * | 1997-11-17 | 2000-11-21 | Kabushiki Kaisha Toshiba | Method of evaluating fault coverage |
US5987636A (en) * | 1998-05-28 | 1999-11-16 | Nec Usa, Inc. | Static test sequence compaction using two-phase restoration and segment manipulation |
US6516432B1 (en) * | 1999-12-22 | 2003-02-04 | International Business Machines Corporation | AC scan diagnostic method |
US6487688B1 (en) * | 1999-12-23 | 2002-11-26 | Logicvision, Inc. | Method for testing circuits with tri-state drivers and circuit for use therewith |
US6718524B1 (en) * | 2001-09-17 | 2004-04-06 | Lsi Logic Corporation | Method and apparatus for estimating state-dependent gate leakage in an integrated circuit |
Cited By (36)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030069724A1 (en) * | 1999-11-30 | 2003-04-10 | Bridges2Silicon, Inc. | Method and system for debugging an electronic system using instrumentation circuitry and a logic analyzer |
US20030131325A1 (en) * | 1999-11-30 | 2003-07-10 | Bridges2Silicon, Inc. | Method and user interface for debugging an electronic system |
US6823497B2 (en) | 1999-11-30 | 2004-11-23 | Synplicity, Inc. | Method and user interface for debugging an electronic system |
US20050010880A1 (en) * | 1999-11-30 | 2005-01-13 | Bridges2Silicon, Inc. | Method and user interface for debugging an electronic system |
US7356786B2 (en) | 1999-11-30 | 2008-04-08 | Synplicity, Inc. | Method and user interface for debugging an electronic system |
US20030115522A1 (en) * | 2001-12-17 | 2003-06-19 | Benoit Nadeau-Dostie | Method and program product for designing hierarchical circuit for quiescent current testing |
US6862717B2 (en) * | 2001-12-17 | 2005-03-01 | Logicvision, Inc. | Method and program product for designing hierarchical circuit for quiescent current testing |
US6769101B2 (en) * | 2002-05-13 | 2004-07-27 | Agilent Technologies, Inc. | Systems and methods providing scan-based delay test generation |
US7827510B1 (en) | 2002-06-07 | 2010-11-02 | Synopsys, Inc. | Enhanced hardware debugging with embedded FPGAS in a hardware description language |
US20040187060A1 (en) * | 2003-03-21 | 2004-09-23 | Rohrbaugh John G. | Generating test patterns for testing an integrated circuit |
US20060259838A1 (en) * | 2005-05-13 | 2006-11-16 | Lewis Nardini | Scan Sequenced Power-On Initialization |
US7469372B2 (en) * | 2005-05-13 | 2008-12-23 | Texas Instruments Incorporated | Scan sequenced power-on initialization |
WO2006138488A3 (en) * | 2005-06-15 | 2008-07-17 | Mentor Graphics Corp | Reduced-pin-count-testing architectures for applying test patterns |
US7487419B2 (en) | 2005-06-15 | 2009-02-03 | Nilanjan Mukherjee | Reduced-pin-count-testing architectures for applying test patterns |
WO2006138488A2 (en) * | 2005-06-15 | 2006-12-28 | Mentor Graphics Corporation | Reduced-pin-count-testing architectures for applying test patterns |
US7308656B1 (en) * | 2005-10-04 | 2007-12-11 | Xilinx, Inc. | Method and apparatus for generating a boundary scan description and model |
US11879941B1 (en) | 2006-02-03 | 2024-01-23 | Texas Instruments Incorporated | Scan testing using scan frames with embedded commands |
US11740286B2 (en) | 2006-02-03 | 2023-08-29 | Texas Instruments Incorporated | Scan testing using scan frames with embedded commands |
US11137447B2 (en) | 2006-02-03 | 2021-10-05 | Texas Instruments Incorporated | IC test architecture having differential data input and output buffers |
US20110314348A1 (en) * | 2006-02-03 | 2011-12-22 | Texas Instruments Incorporated | Scan testing using scan frames with embedded commands |
US10591542B2 (en) | 2006-02-03 | 2020-03-17 | Texas Instruments Incorporated | Input shift register having parallel serial scan outputs, command output |
US8283665B2 (en) * | 2006-02-03 | 2012-10-09 | Texas Instruments Incorporated | Scan paths, stimulus, and header circuitry with command/frame marker outputs |
US10088527B2 (en) | 2006-02-03 | 2018-10-02 | Texas Instruments Incorporated | Differential I/O for parallel scan paths, scan frames, embedded commands |
US9714980B2 (en) | 2006-02-03 | 2017-07-25 | Texas Instruments Incorporated | Scan testing scan frames with embedded commands and differential signaling |
US8618542B2 (en) * | 2006-02-03 | 2013-12-31 | Texas Instruments Incorporated | Response and header circuitry with command and frame marker inputs |
KR101602506B1 (en) | 2009-07-28 | 2016-03-11 | 시놉시스, 인크. | Hierarchical order ranked simulation of electronic circuits |
TWI479349B (en) * | 2009-07-28 | 2015-04-01 | Synopsys Inc | Method of simulating integrated circuit and computer-readable medium |
US8463587B2 (en) | 2009-07-28 | 2013-06-11 | Synopsys, Inc. | Hierarchical order ranked simulation of electronic circuits |
KR20120047336A (en) * | 2009-07-28 | 2012-05-14 | 시놉시스, 인크. | Hierarchical order ranked simulation of electronic circuits |
US20110029299A1 (en) * | 2009-07-28 | 2011-02-03 | Synopsys, Inc. | Hierarchical Order Ranked Simulation Of Electronic Circuits |
WO2011014203A1 (en) * | 2009-07-28 | 2011-02-03 | Synopsys, Inc. | Hierarchical order ranked simulation of electronic circuits |
US8381050B2 (en) | 2009-11-25 | 2013-02-19 | International Business Machines Corporation | Method and apparatus for increased effectiveness of delay and transition fault testing |
US20110121838A1 (en) * | 2009-11-25 | 2011-05-26 | International Business Machines Corporation | Method and apparatus for increased effectiveness of delay and transistion fault testing |
US20150043291A1 (en) * | 2013-08-09 | 2015-02-12 | SK Hynix Inc. | Method for testing semiconductor apparatus and test system using the same |
US9418757B2 (en) * | 2013-08-09 | 2016-08-16 | SK Hynix Inc. | Method for testing semiconductor apparatus and test system using the same |
WO2023070301A1 (en) * | 2021-10-26 | 2023-05-04 | 华为技术有限公司 | Method, apparatus and device for logic simulation |
Also Published As
Publication number | Publication date |
---|---|
WO2003038450A2 (en) | 2003-05-08 |
US6868532B2 (en) | 2005-03-15 |
WO2003038450A3 (en) | 2004-04-22 |
CA2360291A1 (en) | 2003-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6868532B2 (en) | Method and program product for designing hierarchical circuit for quiescent current testing and circuit produced thereby | |
US7058869B2 (en) | Method and apparatus for debug, diagnosis, and yield improvement of scan-based integrated circuits | |
US6003150A (en) | Method for testing field programmable gate arrays | |
US6141790A (en) | Instructions signature and primary input and primary output extraction within an IEEE 1149.1 compliance checker | |
Crouch | Design-for-test for Digital IC's and Embedded Core Systems | |
Murray et al. | Testing ICs: Getting to the core of the problem | |
US7139955B2 (en) | Hierarchically-controlled automatic test pattern generation | |
US6862717B2 (en) | Method and program product for designing hierarchical circuit for quiescent current testing | |
US6012155A (en) | Method and system for performing automatic extraction and compliance checking of an IEEE 1149.1 standard design within a netlist | |
US7562321B2 (en) | Method and apparatus for structured ASIC test point insertion | |
EP0568132A2 (en) | Test generation by environment emulation | |
US5517637A (en) | Method for testing a test architecture within a circuit | |
US7124342B2 (en) | Smart capture for ATPG (automatic test pattern generation) and fault simulation of scan-based integrated circuits | |
US7039845B2 (en) | Method and apparatus for deriving a bounded set of path delay test patterns covering all transition faults | |
Beenker et al. | Testability Concepts for Digital ICs: The Macro Test Approach | |
US20040153926A1 (en) | Method and apparatus for testing asynchronous set/reset faults in a scan-based integrated circuit | |
US7900112B2 (en) | System and method for digital logic testing | |
US11092645B2 (en) | Chain testing and diagnosis using two-dimensional scan architecture | |
KR100683436B1 (en) | Generator of memory BIST circuit | |
Koenemann | Design-for-test | |
US7188043B1 (en) | Boundary scan analysis | |
US20170010320A1 (en) | Reducing test time and system-on-chip (soc) area reduction using simultaneous clock capture based on voltage sensor input | |
Parekhji | Embedded cores and system-on-chip testing | |
Praveen et al. | A Novel Techniques to Reduce the Test Data Volume and Test Time for General Configurations by Embedded Deterministic Testing | |
Chakravarthi | SoC Design for Testability (DFT) |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: LOGICVISION, INC., CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:NADEAU-DOSTIE, BENOIT;COTE, JEAN-FRANCOIS;REEL/FRAME:012373/0189 Effective date: 20011108 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
FEPP | Fee payment procedure |
Free format text: PAT HOLDER NO LONGER CLAIMS SMALL ENTITY STATUS, ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: STOL); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
AS | Assignment |
Owner name: COMERICA BANK, MICHIGAN Free format text: SECURITY AGREEMENT;ASSIGNOR:LOGICVISION, INC.;REEL/FRAME:022629/0938 Effective date: 20090424 |
|
AS | Assignment |
Owner name: LOGICVISION, INC., CALIFORNIA Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:COMERICA BANK;REEL/FRAME:023234/0037 Effective date: 20090911 |
|
FEPP | Fee payment procedure |
Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY Free format text: PAYER NUMBER DE-ASSIGNED (ORIGINAL EVENT CODE: RMPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
AS | Assignment |
Owner name: MENTOR GRAPHICS CORPORATION,OREGON Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:LOGICVISION, INC.;REEL/FRAME:024233/0056 Effective date: 20090910 Owner name: MENTOR GRAPHICS CORPORATION, OREGON Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:LOGICVISION, INC.;REEL/FRAME:024233/0056 Effective date: 20090910 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
FPAY | Fee payment |
Year of fee payment: 12 |
|
AS | Assignment |
Owner name: SIEMENS INDUSTRY SOFTWARE INC., TEXAS Free format text: MERGER AND CHANGE OF NAME;ASSIGNORS:MENTOR GRAPHICS CORPORATION;SIEMENS INDUSTRY SOFTWARE INC.;REEL/FRAME:057279/0707 Effective date: 20201230 |